第一部分

基于iCBCT的宫颈癌放疗分次间器官活动度临床研究

研究目的：基于接受每日迭代锥形束计算机断层成像（iCBCT）引导放疗的宫颈癌，旨在探讨分次间宫颈和宫体移动度，评估不同外放边界对于靶区的覆盖包绕情况。

材料与方法：本研究分析了15例接受外照射的宫颈癌共303次iCBCT图像，以子宫峡部为界，将临床靶区（CTV）分为两部分，一部分为宫体临床靶区（CTV-U），包括宫体；另一部分为宫颈临床靶区（CTV-C），包括原发肿瘤、宫颈、临近的宫旁以及部分阴道。基于iCBCT图像勾画所有的CTV-U₂和CTV-C₂，通过刚性配准投射到定位CT上面，形成最终的CTV-U₃和CTV-C₃。利用边界算法开发一款软件，将其与定位CT上的初始CTV-U₁和CTV-C₁作比较，测量靶区在不同方向的移动度。同时对定位CT靶区进行均匀外扩，评估覆盖包绕情况。

研究结果：CTV-U活动度（均值±标准差）左侧为8.3±4.1 mm，右侧为9.8±4.4 mm，前侧为12.6±4.0 mm，后侧为8.8±5.1 mm，头侧为5.7±5.4 mm，脚侧为3.0±3.2 mm。CTV-C的活动度左侧为7.3±3.2 mm，右侧为8.6±3.8 mm，前侧为9.0±6.1 mm，后侧为8.4±3.6 mm，头侧为5.0±5.0 mm，脚侧为3.0±2.5 mm。与T2和T3期相比，T1期患者的CTV-U和CTV-C运动范围更大。采用15 mm均匀外放CTV边界，分别有11.1%和2.2%的治疗分次未能完全覆盖CTV-U₃和CTV-C₃。与定位CT的初始靶区CTV-U₁和CTV-C₁相比，CTV-U₃和CTV-C₃的平均体积变化分别为150%和51%。

研究结论：利用边界算法结合每日iCBCT图像，能够有效评估宫颈癌分次间CTV-U和CTV-C的活动度。宫体活动度大于宫颈，早期T分期患者需更大外放边界。个性化外放边界有助于提高靶区剂量并减少正常组织照射，在线自适应放疗可能为未来宫颈癌精准化放疗提供了方向。

第二部分

宫颈癌中等分割在线自适应放疗可行性研究

研究目的：基于在线自适应放疗（oART）技术，从应用流程、靶区、危及器官（OARs）剂量，以及治疗完成度等方面评估宫颈癌中等分割放疗（MHRT）开展的可行性。

材料与方法：前瞻性入组2023年9月至2024年4月期间本中心接受根治性放疗的宫颈癌患者。所有患者均接受基于iCBCT引导的每日oART，CTV处方剂量为43.35Gy/17次，阳性淋巴结则采用同步推量，剂量为54.40Gy/17次。研究采用oART工作流程，具体包括：iCBCT图像采集、人工智能辅助靶区勾画、放疗计划适应性调整及治疗验证等环节。详细记录各流程的时间消耗，并通过靶区和OARs的剂量学指标进行评价，并记录治疗完成率等临床指标。

研究结果：30例患者共完成510个分次oART治疗，首次oART成功实施率达到99%。每次治疗的平均时长为23分18秒，其中自适应流程（从首次iCBCT扫描到治疗计划选择）平均耗时17分20秒。99.4%的分次选择了adapted计划，显著改善了CTV和计划靶区（PTV）的剂量覆盖（V_100%、V_98%、V_95%、V_90%），尤其是宫体和宫颈PTV的V_100%，分别提高了7.26%和8.79%（P< 0.001）。自适应计划中小肠、直肠和膀胱的D_mean、V_10Gy、V_{20 Gy}及V_{40 Gy}等剂量指标也显著降低。此外，自适应计划还改善了骨髓和股骨头的剂量学指标。

研究结论：基于oART，宫颈癌MHRT显示出良好的可行性。未来需进一步探索其预后和毒副反应，为宫颈癌放疗模式的优化提供依据。

第三部分

宫颈癌中等分割在线自适应放疗的疗效及安全性分析：一项前瞻性单臂I期临床研究

研究目的：开展宫颈癌中等分割oART的前瞻性I期临床研究，评估患者初步疗效及急性毒副反应。

材料与方法：前瞻性入组病理确诊的宫颈癌患者30例，所有受试者接受每天oART，CTV处方剂量为43.35Gy/17次，阳性淋巴结则采用同步推量，剂量为54.40Gy/17次。CTV分开勾画并根据不同解剖结构进行5-10 mm外放。CT/MRI引导的三维近距离放疗，处方剂量30Gy/5次，同步顺铂40mg/m²周疗，至少3周期。肿瘤疗效和毒副反应分别通过RECIST 1.1和CTCAE 5.0标准评价分级，通过欧洲癌症研究与治疗组织生命质量核心量表（EORTC-QLQ-C30）和欧洲癌症研究与治疗组织宫颈癌特异性生命质量量表（EORTC-QLQ-CX24）量表报道患者不同治疗时间点的生活质量。

研究结果：30例患者完成中等分割oART联合近距离治疗，外照射中位时间为22天（范围：22-27天），总治疗中位时间为43天（范围：38-55天）。外照射完成时肿瘤平均退缩率为83.27%，3个月的临床完全缓解率（CCR）为100%。中位随访13.0个月，1年的总生存率（OS）率和无病生存率（DFS）分别为100%和93.2%。急性毒性反应中，3级胃肠道毒性发生率为10%，无3级及以上泌尿系毒性。生活质量方面，中等分割oART期间在功能领域得分显著降低，在症状领域得分显著升高，在近距离治疗期间开始改善，治疗后1个月和4个月逐步恢复基线水平甚至更高。

研究结论：宫颈癌MHRT联合近距离治疗具有缩短治疗时间的优势，且疗效和毒性反应可接受。然而，仍需长期随访和大样本多中心研究进一步验证其应用前景。

第四部分

宫颈癌中等分割放疗与常规分割放疗临床应用的比较：一项倾向性评分匹配分析

研究目的：比较宫颈癌常规分割放疗（CFRT）和MHRT的疗效及安全性，为前瞻性随机对照研究提供证据。

材料与方法：本研究纳入FIGO 2018分期为IB-IIB和IIIC1期（≤1.5 cm转移淋巴结、无髂淋巴结转移且盆腔淋巴结转移≤2枚）的宫颈鳞癌患者。入组患者分为两组：一组为基于每天oART的MHRT，处方剂量为43.35Gy/17次，阳性淋巴结同步推量至54.40Gy/17次；另一组为基于每天图像引导的CFRT，处方剂量为50.4Gy/28次，阳性淋巴结同步推量至60.2Gy/28次。研究的主要终点为急性毒性反应，次要终点包括外照射后肿瘤退缩情况、CCR、OS、DFS及局部控制率（LC）。并对倾向性评分匹配（PSM）前后的疗效和安全性进行对比分析。

研究结果：本研究纳入了2023年9月至2024年4月期间接受根治性放疗的宫颈癌192例，其中162例接受CFRT，30例接受MHRT。在PSM后，两组患者的临床特征达到平衡。MHRT组治疗时间显著缩短，外照射时间从CFRT组的42天缩短至22天（P< 0.001），总治疗时间从57天缩短至43天（P< 0.001）。在肿瘤退缩方面，PSM前后MHRT组每分次和每天的肿瘤退缩率均显著高于CFRT组。3个月的CCR相似（96.2% vs. 100%，P= 0.150），1年的OS率、DFS率、LC率无明显统计学差异。在安全性方面，PSM前CFRT组任何2级及以上的急性毒性发生率较高（98.1% vs. 90%，P= 0.018）；PSM后两组急性毒性发生率相近，≥3级急性毒性发生率无显著差异。PSM前MHRT组3级及以上胃肠道毒性发生率较高（P= 0.028），但匹配后差异有所减小。两组均未观察到3级及以上泌尿系统毒性，且PSM前后血液学毒性发生率相似。

研究结论：MHRT在宫颈癌治疗中表现出与CFRT相当的近期疗效和安全性，且治疗时间更短，肿瘤退缩率更高。MHRT有望成为CFRT的替代方案。

Part I

Pelvic target volume inter-fractional motion during radiotherapy for cervical cancer with daily iterative cone beam computed tomography.

Objective: Based on daily iterative cone beam CT (iCBCT)-guided radiotherapy in cervical cancer patients, this study aimed to quantify inter-fractional movements of the uterus and cervix and to evaluate the clinical target volume (CTV) coverage.

Methods: This study analyzed 303 iCBCT scans from 15 cervical cancer patients undergoing external beam radiotherapy. The cervix and uterus were also contoured separately at the uterine isthmus. The clinical target volume of the uterus (CTV-U) included the uterus, while the clinical target volume of the cervix (CTV-C) included the tumor, cervix, vagina, and adjacent parametrial regions. All CTV-U2 and CTV-C2 contours delineated on iCBCT images were projected onto the planning CT through rigid registration, forming the final CTV-U3 and CTV-C3. Software was developed using boundary algorithms, and these were compared with the initial CTV-U1 and CTV-C1 on the planning CT to measure target displacement in different directions. Simultaneously, uniform expansions were applied to the planning CTV to assess target coverage.

Results: The motion (mean±standard deviation) in the CTV-U position was 8.3±4.1 mm in the left, 9.8±4.4 mm in the right, 12.6±4.0 mm in the anterior, 8.8±5.1 mm in the posterior, 5.7±5.4 mm in the superior, and 3.0±3.2 mm in the inferior direction. The mean CTV-C displacement was 7.3±3.2 mm in the left, 8.6±3.8 mm in the right, 9.0±6.1 mm in the anterior, 8.4±3.6 mm in the posterior, 5.0±5.0 mm in the superior, and 3.0±2.5 mm in the inferior direction. Compared with the stage T2 and stage T3, CTV-U and CTV-C motion in stage T1 was larger. A uniform CTV planning treatment volume margin of 15 mm failed to encompass the CTV-U3 and CTV-C3 in 11.1% and 2.2% of all fractions, respectively. The mean volume changes of CTV-U and CTV-C were 150% and 51%, respectively, compared with the planning CTV.

Conclusion: The application of margin algorithms combined with daily iCBCT images enables effective assessment of intra-fractional motion for CTV-U and CTV-C in cervical cancer. Movements of the uterine corpus are larger than those of the cervix, necessitating larger margins for early T-stage patients. Personalized margin strategies can enhance target dose coverage while reducing normal tissue irradiation. Online adaptive radiotherapy may provide a promising direction for precision radiotherapy in cervical cancer treatment.

Part II

Feasibility of moderately hypofractionated online adaptive radiotherapy in the treatment of cervical cancer

Objective: This study evaluates the feasibility of implementing moderate hypofractionated radiotherapy (MHRT) for cervical cancer using online adaptive radiotherapy (oART) technology, with comprehensive assessment of workflow, target volume coverage, organ-at-risk (OARs) dosimetry, and treatment completion rates.

Methods: This prospective study enrolled cervical cancer patients scheduled for definitive radiotherapy between September 2023 and April 2024. All patients underwent daily iterative cone beam computed tomography (iCBCT)-guided oART, with a prescribed dose of 43.35Gy/17 fractions to clinical target volume (CTV) and simultaneous integrated boost of 54.40Gy/17 fractions to positive lymph nodes. The study employed an online adaptive workflow, involving iCBCT scans for image acquisition, artificial intelligence-assisted contouring, plan adaptation, and treatment verification. Time for each process was tracked, and dosimetric assessments were performed to evaluate target volume coverage and OARs protection. Treatment completion rate was reported.

Results: The treatment was successfully implemented for 510 sessions of oART in 30 patients, with a 99.0% success rate for the first oART session. The average time required for each treatment fraction was 23 minutes and 18 seconds, with the adaptive workflow (from the first iCBCT scan to selecting the adapted plan) taking an average of 17 minutes and 20 seconds. The adapted plan was selected in 99.4% of fractions and significantly improved the dosimetric coverage (V_100%, V_98%, V_95%, V_90%) for CTV and planning target volume (PTV), especially V_100% for PTV of the uterus and PTV of the cervix, which improved by nearly 7.26% and 8.79%, respectively (P< 0.001). In the adaptive plan, dose parameters including D_mean, V_10Gy, V_20Gy, and V_40Gy for the small intestine, rectum, and bladder were also significantly reduced. The dosimetry for the bone marrow and femoral heads was also improved with adaptation.

Conclusion: The implementation of MHRT for cervical cancer using oART demonstrates promising feasibility. Future studies should focus on investigating long-term outcomes and toxicity profiles, which may provide valuable evidence for optimizing radiotherapy paradigms in cervical cancer treatment.

Part III

Efficacy and safety of moderate hypofractionated online adaptive radiotherapy for cervical cancer: a prospective phase I clinical study

Objective: To conduct a prospective Phase I clinical trial evaluating the preliminary efficacy and acute toxicities of moderate hypofractionated online adaptive radiotherapy (oART) for cervical cancer.

Methods: Thirty patients with pathologically confirmed cervical cancer were prospectively enrolled. All participants received daily oART, with a prescribed dose of 43.35 Gy in 17 fractions to the clinical target volume (CTV) and a simultaneous integrated boost of 54.40 Gy in 17 fractions to the positive lymph nodes. The CTV was delineated separately with 5-10 mm margins based on anatomical structures. CT/MRI-guided three-dimensional brachytherapy was administered with a prescribed dose of 30 Gy in 5 fractions, concurrent with weekly cisplatin at 40 mg/m² for at least 3 cycles. Tumor response and acute toxicities were assessed using RECIST 1.1 and CTCAE 5.0 criteria, respectively. Quality of life was evaluated at different treatment time points using the EORTC-QLQ-C30 and EORTC-QLQ-CX24 questionnaires.

Results: All 30 patients completed moderate hypofractionated oART combined with brachytherapy. The median duration of external beam radiotherapy was 22 days (range: 22-27 days), and the total median treatment time was 43 days (range: 38-55 days). The average tumor regression rate during external beam radiotherapy was 83.27%, with a 3-month complete clinical response rate of 100%. At a median follow-up of 13.0 months, the 1-year overall survival and disease-free survival rates were 100% and 93.3%, respectively. Acute toxicities included grade 3 gastrointestinal toxicity in 10% of patients, with no grade 3 or higher genitourinary toxicities. Quality of life analysis revealed significant declines in functional domain scores and increases in symptom domain scores during oART, with improvement beginning during brachytherapy and returning to or exceeding baseline levels at 1 and 4 months post-treatment.

Conclusion: Moderate hypofractionated oART combined with brachytherapy offers the advantage of shortened treatment duration, with acceptable efficacy and toxicity profiles. However, long-term follow-up and large-scale multicenter studies are needed to further validate its clinical potential.

Part IV

Comparison of moderate hypofractionated radiotherapy and conventional fractionated radiotherapy in cervical cancer: a propensity score-matched analysis

Objective: To compare the efficacy and safety of conventional fractionated radiotherapy (CFRT) and moderate hypofractionated radiotherapy (MHRT) in cervical cancer, providing evidence for future randomized controlled trials.

Methods: This study included patients with FIGO stage IB-IIB or IIIC1 cervical squamous cell carcinoma, including those with metastatic lymph nodes ≤1.5 cm, no iliac node metastasis, and ≤2 pelvic lymph node metastases. Patients were divided into two groups: one group received daily oART-based MHRT, with a prescribed dose of 43.35 Gy in 17 fractions and a simultaneous integrated boost delivering up to 54.40 Gy to positive lymph nodes. The other group received daily image-guided CFRT, with a prescribed dose of 50.4 Gy in 28 fractions and a simultaneous integrated boost delivering up to 60.2 Gy to positive lymph nodes. The primary endpoint was acute toxicity, while secondary endpoints included tumor regression after external beam radiotherapy (EBRT), complete clinical response (CCR) rate, overall survival (OS), disease-free survival (DFS), and local control (LC) rate. Efficacy and safety were compared before and after propensity score matching (PSM).

Results: A total of 192 patients treated between September 2023 and April 2024 were included: 162 received CFRT, and 30 received MHRT. After PSM, the clinical characteristics were balanced between the groups. The MHRT group had significantly shorter treatment durations, with EBRT taking 22 days versus 42 days for CFRT (P< 0.001) and total treatment taking 43 days versus 57 days (P< 0.001). The MHRT group showed significantly higher tumor shrinkage per session and per day before and after PSM. At 3 months, CCR were similar (96.2% vs. 100%, P= 0.150), and survival outcomes (OS, DFS, LC) showed no significant differences at 1 year. In terms of safety, the incidence of any acute grade 2 or higher toxicities was higher in the CFRT group (98.1% vs. 90%, P= 0.018) before PSM. After PSM, both groups had similar rates of any acute toxicities, with no significant differences in grade 3 or higher adverse events. Acute grade 3 or higher gastrointestinal toxicity was more common in the MHRT group before PSM (P=0.028), but this was less pronounced after matching. No grade 3 or higher urinary toxicity was observed in either group. Hematological toxicities were similar between groups before and after PSM.

Conclusion: MHRT demonstrated comparable short-term efficacy and safety to CFRT in the treatment of cervical cancer, with the added benefit of shorter treatment durations and higher tumor regression rates. MHRT shows promise as a potential alternative to CFRT.